Fix OTTO Schema Deployment Errors and Failures

Camilo Aponte

Camilo Aponte

Last updated on Sep 30, 2026

Overview

When OTTO generates schemas automatically, the AI can occasionally produce invalid data — such as numeric fields stored as text, corrupted characters (like malformed ampersands), or null-valued fields. These errors cause schemas to fail validation and become stuck in an undeployed state, and you may see a message like No deployable recommendation found.

This article explains the known issue and provides concrete steps you can take to resolve it. In cases where the malformed data cannot be corrected through the steps below, the Search Atlas support team can investigate and resolve the affected recommendations directly.

Common Symptoms

  • Attempting to deploy a schema has no effect.
  • A message appears saying No deployable recommendation found.
  • Schemas remain stuck in an undeployed state even after repeated attempts.
  • Schemas were generated for product pages that have no review or rating data, resulting in empty or invalid fields.

What To Do

  1. Identify the affected schemas. Navigate to the OTTO section of the platform and locate the schemas that are failing to deploy. Note which recommendations are sitting in an undeployed or pending state.
  2. Check for obvious data issues. Look at the field values in the affected schema recommendations. Common problems include: numeric fields (such as width, height, price, or rating values) appearing as text strings rather than numbers; special characters that look corrupted or broken (such as malformed ampersands in titles or descriptions); and fields that are empty or contain null values where real content is expected.
  3. Edit the affected schema fields manually. If you can identify the malformed field, edit it directly within the schema recommendation to correct the value — for example, change a text-formatted number to a proper numeric value, remove corrupted characters, or delete empty fields that are causing validation failures. Save the changes and attempt to deploy the schema again.
  4. Remove invalid review or rating fields for pages with no review data. If a schema was generated for a page that has no reviews or ratings, open the schema recommendation and remove the review and rating fields entirely. Schemas without those fields can deploy successfully when the source page does not contain that data. Save and attempt deployment again.
  5. Attempt deployment after corrections. After editing the affected fields, try deploying the schema again. In most cases, correcting the malformed or empty fields resolves the validation error and allows the schema to deploy successfully.
  6. Contact Search Atlas support if schemas remain stuck. If you have corrected the visible field issues and the schema still cannot be deployed, reach out to the support team with the details of the affected schemas and account. The team can investigate the underlying data at the account level and resolve any remaining malformed recommendations directly.

Why This Happens

The OTTO AI schema generator can produce field values in the wrong format or leave required fields empty when source page data is missing or ambiguous. This results in schemas that fail internal validation and cannot be deployed until the underlying data problems are corrected.

If you need further assistance, please contact the Search Atlas support team through the chat widget or your account's support channel and we'll be happy to help.